Behind the growing interest in Generate AI and LLM-based enterprise applications lies an expanded set of requirements for data integrations and ML orchestration. Enterprises want to use proprietary data to power LLM-based applications that create new business value, but they face challenges in moving beyond experimentation. The pipelines that power these models need to run reliably at scale, bringing together data from many sources and reacting continuously to changing conditions. This talk focuses on the design patterns for using Apache Airflow to support LLM applications created using private enterprise data. We’ll go through a real-world example of what this looks like, as well as a proposal to improve Airflow and to add additional Airflow Providers to make it easier to interact with LLMs such as the ones from OpenAI (such as GPT4) and the ones on HuggingFace, while working with both structured and unstructured data. In short, this shows how these Airflow patterns enable reliable, traceable, and scalable LLM applications within the enterprise.

New users starting with Airflow frequently encounter several challenges, ranging from the complexities of Containers and virtual environments to the Python dependency hell. Moreover, their familiarity with tools such as Docker, docker-compose, and Helm might be somewhat limited and even overkill. In contrast, seasoned Airflow users encounter their problems, encompassing configuration conflics with ongoing Airflow projects and intricacies stemming from Docker and docker-compose configurations and lack of visibility into all the projects. With airflowctl, users can install & setup Airflow using a single command. For existing users, they can use it to manage multiple Airflow projects with different Airflow versions on the same machine. This allows creating & debugging DAGs in an IDE seamlessly. Agenda for the call: Why airflowctl? Goal Current functionality & Demo Vision / Roadmap